[zattoo] Improve and simplify (closes #14676)

youtube_dl/extractor/zattoo.py

@@ -1,84 +1,82 @@
 # coding: utf-8
 from __future__ import unicode_literals
 
-from uuid import uuid4
 import re
+from uuid import uuid4
 
 from .common import InfoExtractor
-from ..utils import (
+from ..compat import (
+    compat_HTTPError,
     compat_str,
+)
+from ..utils import (
     ExtractorError,
-    sanitized_Request,
+    int_or_none,
+    try_get,
     urlencode_postdata,
 )
 
 
 class ZattooBaseIE(InfoExtractor):
-
     _NETRC_MACHINE = 'zattoo'
     _HOST_URL = 'https://zattoo.com'
 
     _power_guide_hash = None
 
-    def _login(self, uuid, session_id):
+    def _login(self):
         (username, password) = self._get_login_info()
         if not username or not password:
-            raise ExtractorError(
-                'A valid %s account is needed to access this media.' % self._NETRC_MACHINE,
-                expected=True)
-        login_form = {
-            'login': username,
-            'password': password,
-            'remember': True,
-        }
-        request = sanitized_Request(
-            '%s/zapi/v2/account/login' % self._HOST_URL,
-            urlencode_postdata(login_form))
-        request.add_header(
-            'Referer', '%s/login' % self._HOST_URL)
-        request.add_header(
-            'Content-Type', 'application/x-www-form-urlencoded; charset=UTF-8')
-        request.add_header(
-            'Cookie', 'uuid=%s; beaker.session.id=%s' % (uuid, session_id))
-        response = self._request_webpage(
-            request, None, 'Logging in')
-        data = self._parse_json(response.read(), None)
-        return data['session']['power_guide_hash']
-
-    def _get_app_token_and_version(self):
-        host_webpage = self._download_webpage(
-            self._HOST_URL, None, 'Downloading %s' % self._HOST_URL)
+            self.raise_login_required(
+                'A valid %s account is needed to access this media.'
+                % self._NETRC_MACHINE)
+
+        try:
+            data = self._download_json(
+                '%s/zapi/v2/account/login' % self._HOST_URL, None, 'Logging in',
+                data=urlencode_postdata({
+                    'login': username,
+                    'password': password,
+                    'remember': 'true',
+                }), headers={
+                    'Referer': '%s/login' % self._HOST_URL,
+                    'Content-Type': 'application/x-www-form-urlencoded; charset=UTF-8',
+                })
+        except ExtractorError as e:
+            if isinstance(e.cause, compat_HTTPError) and e.cause.code == 400:
+                raise ExtractorError(
+                    'Unable to login: incorrect username and/or password',
+                    expected=True)
+            raise
+
+        self._power_guide_hash = data['session']['power_guide_hash']
+
+    def _real_initialize(self):
+        webpage = self._download_webpage(
+            self._HOST_URL, None, 'Downloading app token')
         app_token = self._html_search_regex(
-            r'<script.+window\.appToken\s*=\s*\'(.+)\'', host_webpage, 'app token')
+            r'appToken\s*=\s*(["\'])(?P<token>(?:(?!\1).)+?)\1',
+            webpage, 'app token', group='token')
         app_version = self._html_search_regex(
-            r'<!--\w+-(.+?)-', host_webpage, 'app version', default='2.8.2')
-        return app_token, app_version
-
-    def _say_hello(self, uuid, app_token, app_version):
-        postdata = {
-            'client_app_token': app_token,
-            'uuid': uuid,
-            'lang': 'en',
-            'app_version': app_version,
-            'format': 'json',
-        }
-        request = sanitized_Request(
-            '%s/zapi/v2/session/hello' % self._HOST_URL,
-            urlencode_postdata(postdata))
-        response = self._request_webpage(
-            request, None, 'Say hello')
+            r'<!--\w+-(.+?)-', webpage, 'app version', default='2.8.2')
+
+        # Will setup appropriate cookies
+        self._request_webpage(
+            '%s/zapi/v2/session/hello' % self._HOST_URL, None,
+            'Opening session', data=urlencode_postdata({
+                'client_app_token': app_token,
+                'uuid': compat_str(uuid4()),
+                'lang': 'en',
+                'app_version': app_version,
+                'format': 'json',
+            }))
 
-        cookie = response.headers.get('Set-Cookie')
-        session_id = self._search_regex(
-            r'beaker\.session\.id\s*=\s*(.+?);', cookie, 'session id')
-        return session_id
+        self._login()
 
     def _extract_cid(self, video_id, channel_name):
         channel_groups = self._download_json(
             '%s/zapi/v2/cached/channels/%s' % (self._HOST_URL,
                                                self._power_guide_hash),
-            video_id,
-            'Downloading available channel list',
+            video_id, 'Downloading channel list',
             query={'details': False})['channel_groups']
         channel_list = []
         for chgrp in channel_groups:
@@ -86,7 +84,9 @@ class ZattooBaseIE(InfoExtractor):
         try:
             return next(
                 chan['cid'] for chan in channel_list
-                if chan['display_alias'] == channel_name or chan['cid'] == channel_name)
+                if chan.get('cid') and (
+                    chan.get('display_alias') == channel_name or
+                    chan.get('cid') == channel_name))
         except StopIteration:
             raise ExtractorError('Could not extract channel id')
 
@@ -100,72 +100,90 @@ class ZattooBaseIE(InfoExtractor):
                 'complete': True
             })
 
+        p = data['program']
+        cid = p['cid']
+
         info_dict = {
             'id': video_id,
-            'title': data['program']['title'],
-            'description': data['program'].get('description'),
-            'thumbnail': data['program'].get('image_url')
+            'title': p.get('title') or p['episode_title'],
+            'description': p.get('description'),
+            'thumbnail': p.get('image_url'),
+            'creator': p.get('channel_name'),
+            'episode': p.get('episode_title'),
+            'episode_number': int_or_none(p.get('episode_number')),
+            'season_number': int_or_none(p.get('season_number')),
+            'release_year': int_or_none(p.get('year')),
+            'categories': try_get(p, lambda x: x['categories'], list),
         }
-        cid = data['program']['cid']
+
         return cid, info_dict
 
     def _extract_formats(self, cid, video_id, record_id=None, is_live=False):
-        postdata = {
-            'stream_type': 'dash',
+        postdata_common = {
             'https_watch_urls': True,
         }
-        if record_id:
-            url = '%s/zapi/watch/recording/%s' % (self._HOST_URL, record_id)
-        else:
-            url = '%s/zapi/watch/recall/%s/%s' % (self._HOST_URL, cid, video_id)
 
         if is_live:
-            postdata.update({'timeshift': 10800})
+            postdata_common.update({'timeshift': 10800})
             url = '%s/zapi/watch/live/%s' % (self._HOST_URL, cid)
-
-        data = self._download_json(
-            sanitized_Request(url, urlencode_postdata(postdata)),
-            video_id, 'Downloading dash formats')
+        elif record_id:
+            url = '%s/zapi/watch/recording/%s' % (self._HOST_URL, record_id)
+        else:
+            url = '%s/zapi/watch/recall/%s/%s' % (self._HOST_URL, cid, video_id)
 
         formats = []
-        for elem in data['stream']['watch_urls']:
-            audio_channel = elem.get('audio_channel')
-            maxrate = elem.get('maxrate')
-            formats.extend(
-                self._extract_mpd_formats(
-                    elem['url'], video_id,
-                    mpd_id='dash-maxrate-%s-channel-%s' % (maxrate, audio_channel), fatal=False))
-
-        postdata.update({'stream_type': 'hls'})
-        request = sanitized_Request(
-            url, urlencode_postdata(postdata))
-        data = self._download_json(
-            request, video_id, 'Downloading hls formats')
-        for elem in data['stream']['watch_urls']:
-            audio_channel = elem.get('audio_channel')
-            preference = None
-
-            # Prefer audio channel A:
-            if audio_channel == 'A':
-                preference = 1
-
-            maxrate = elem.get('maxrate')
-            formats.extend(
-                self._extract_m3u8_formats(
-                    elem['url'], video_id, 'mp4', entry_protocol='m3u8_native',
-                    preference=preference,
-                    m3u8_id='hls-maxrate-%s-channel-%s' % (maxrate, audio_channel),
-                    fatal=False))
-
+        for stream_type in ('dash', 'hls', 'hls5', 'hds'):
+            postdata = postdata_common.copy()
+            postdata['stream_type'] = stream_type
+
+            data = self._download_json(
+                url, video_id, 'Downloading %s formats' % stream_type.upper(),
+                data=urlencode_postdata(postdata), fatal=False)
+            if not data:
+                continue
+
+            watch_urls = try_get(
+                data, lambda x: x['stream']['watch_urls'], list)
+            if not watch_urls:
+                continue
+
+            for watch in watch_urls:
+                if not isinstance(watch, dict):
+                    continue
+                watch_url = watch.get('url')
+                if not watch_url or not isinstance(watch_url, compat_str):
+                    continue
+                format_id_list = [stream_type]
+                maxrate = watch.get('maxrate')
+                if maxrate:
+                    format_id_list.append(compat_str(maxrate))
+                audio_channel = watch.get('audio_channel')
+                if audio_channel:
+                    format_id_list.append(compat_str(audio_channel))
+                preference = 1 if audio_channel == 'A' else None
+                format_id = '-'.join(format_id_list)
+                if stream_type in ('dash', 'dash_widevine', 'dash_playready'):
+                    this_formats = self._extract_mpd_formats(
+                        watch_url, video_id, mpd_id=format_id, fatal=False)
+                elif stream_type in ('hls', 'hls5', 'hls5_fairplay'):
+                    this_formats = self._extract_m3u8_formats(
+                        watch_url, video_id, 'mp4',
+                        entry_protocol='m3u8_native', m3u8_id=format_id,
+                        fatal=False)
+                elif stream_type == 'hds':
+                    this_formats = self._extract_f4m_formats(
+                        watch_url, video_id, f4m_id=format_id, fatal=False)
+                elif stream_type == 'smooth_playready':
+                    this_formats = self._extract_ism_formats(
+                        watch_url, video_id, ism_id=format_id, fatal=False)
+                else:
+                    assert False
+                for this_format in this_formats:
+                    this_format['preference'] = preference
+                formats.extend(this_formats)
         self._sort_formats(formats)
         return formats
 
-    def _real_initialize(self):
-        uuid = compat_str(uuid4())
-        app_token, app_version = self._get_app_token_and_version()
-        session_id = self._say_hello(uuid, app_token, app_version)
-        self._power_guide_hash = self._login(uuid, session_id)
-
     def _extract_video(self, channel_name, video_id, record_id=None, is_live=False):
         if is_live:
             cid = self._extract_cid(video_id, channel_name)
@@ -190,13 +208,27 @@ class QuicklineBaseIE(ZattooBaseIE):
 class QuicklineIE(QuicklineBaseIE):
     _VALID_URL = r'https?://(?:www\.)?mobiltv\.quickline\.com/watch/(?P<channel>[^/]+)/(?P<id>[0-9]+)'
 
+    _TEST = {
+        'url': 'https://mobiltv.quickline.com/watch/prosieben/130671867-maze-runner-die-auserwaehlten-in-der-brandwueste',
+        'only_matching': True,
+    }
+
     def _real_extract(self, url):
         channel_name, video_id = re.match(self._VALID_URL, url).groups()
         return self._extract_video(channel_name, video_id)
 
 
 class QuicklineLiveIE(QuicklineBaseIE):
-    _VALID_URL = r'https?://(?:www\.)?mobiltv\.quickline\.com/watch/(?P<id>[^/]+)$'
+    _VALID_URL = r'https?://(?:www\.)?mobiltv\.quickline\.com/watch/(?P<id>[^/]+)'
+
+    _TEST = {
+        'url': 'https://mobiltv.quickline.com/watch/srf1',
+        'only_matching': True,
+    }
+
+    @classmethod
+    def suitable(cls, url):
+        return False if QuicklineIE.suitable(url) else super(QuicklineLiveIE, cls).suitable(url)
 
     def _real_extract(self, url):
         channel_name = video_id = self._match_id(url)
@@ -222,13 +254,17 @@ class ZattooIE(ZattooBaseIE):
 
 
 class ZattooLiveIE(ZattooBaseIE):
-    _VALID_URL = r'https?://(?:www\.)?zattoo\.com/watch/(?P<id>[^/]+)$'
+    _VALID_URL = r'https?://(?:www\.)?zattoo\.com/watch/(?P<id>[^/]+)'
 
     _TEST = {
         'url': 'https://zattoo.com/watch/srf1',
         'only_matching': True,
     }
 
+    @classmethod
+    def suitable(cls, url):
+        return False if ZattooIE.suitable(url) else super(ZattooLiveIE, cls).suitable(url)
+
     def _real_extract(self, url):
         channel_name = video_id = self._match_id(url)
         return self._extract_video(channel_name, video_id, is_live=True)